• Energy Research

Fault-induced delayed voltage recovery (FIDVR) events have become a critical threat to modern power systems with high-level inverter-interfaced renewable power generation. Aiming at the real-time assessment on FIDVR, this paper proposes a data-driven method using real-time bus voltage trajectory measurements. Based on ensemble learning and probabilistic prediction techniques, a self-adaptive decision-making model is developed to rapidly predict the FIDVR severity index following a disturbance in the system. The salient feature of the proposed method is that the FIDVR assessment result can be delivered as early as possible without impairing the assessment accuracy, thereby more time is available for emergency controls. The proposed method is tested on New England 39-bus system, and the results demonstrate its high accuracy and exceptionally faster speed over existing methods.

Due to the rapid development of distributed renewable generation, an effective risk assessment and early warning mechanism for active distribution networks is of great significance to maintain the system reliability and enhance energy grid resilience. In this paper, a novel risk assessment model is proposed to assess the probability of potential disturbances to the grid and provide accurate advice for trading prosumers’ renewable energy. The model can compute node failure probability (FP) for transmission networks as well as the area FP for distribution networks, while combining the two perspectives by topology analysis. A weather threshold value is first derived to define the extreme weather condition. Then the FP of transmission lines is calculated by joint probability models under four instances of extreme climate. For distribution networks, the weather influence is obtained by applying the Rare Events Logistic Regression model initially. Then the equipment fault related to the geographical feature is captured using feeder taxonomy and hierarchical clustering. Furthermore, the accidental factor as a new parameter is introduced to evaluate the vandalism, vegetation, and operating fault to the grid. Finally, the warning information and advice for customers will be presented after fault chain analysis. The FP for a specific area in Australia is analyzed in case studies to verify the proposed model.

As the power system is facing a transition toward a more intelligent, flexible, and interactive system with higher penetration of renewable energy generation, load forecasting, especially short-term load forecasting for individual electric customers plays an increasingly essential role in the future grid planning and operation. Other than aggregated residential load in a large scale, forecasting an electric load of a single energy user is fairly challenging due to the high volatility and uncertainty involved. In this paper, we propose a long short-term memory (LSTM) recurrent neural network-based framework, which is the latest and one of the most popular techniques of deep learning, to tackle this tricky issue. The proposed framework is tested on a publicly available set of real residential smart meter data, of which the performance is comprehensively compared to various benchmarks including the state-of-the-arts in the field of load forecasting. As a result, the proposed LSTM approach outperforms the other listed rival algorithms in the task of short-term load forecasting for individual residential households.

With the increase of load fluctuations and the integration of stochastic distributed generations (DGs), there have been more and more research interests in forecasting-aided state estimation. In this paper, we propose a robust generalized maximum likelihood (GM)-estimator based power system forecasting-aided state estimation, which integrates the statistical characteristics of both loads and DGs, i.e., spatial and temporal correlations. A first order vector auto-regressive model (VAR(1)) is developed to capture the statistical characteristics of load and DGs, facilitating short-term loads and DGs forecasting. These forecasted power injections are further combined with power balance equations to derive a new state transition model, where the relationship between forecasted state vector and predicted power injections is expressed explicitly. After that, a redundant batch regression model that simultaneously processes predicted state vector and received observations is derived, allowing the development of a robust estimator. To this end, we propose a robust GM-estimator that leverages modified projection statistics and a Huber convex score function, to bound the influence of observation outliers while maintaining its high statistical estimation efficiency. Finally, the iteratively reweighted least squares algorithm is adopted to solve the GM-estimator. Numerical comparisons on IEEE benchmark systems with DGs integration demonstrate the efficiency and robustness of the proposed method.

False data injection attack (FDIA) is a major threat in wide-area monitoring systems. Being able to differentiate FDIA from normal grid contingencies is a paramount necessity for a grid operator to decide the correct response on a critical prompt basis as well as reduce the overall FDIAs false alarms. Two FDIAs characterization algorithms are developed in this paper. The first is based on the principal component analysis (PCA) while the second is based on the canonical correlation analysis (CCA). Both algorithms are developed in an online platform to reduce the computational complexity. The various designed test cases illustrate a promising FDIA characterization performance using both algorithms. The testing results of three machine learning-based classifiers indicate that the proposed FDIAs characterization algorithms provide better classification models than conventional PCA-based characterization algorithm with CCA illustrating advanced characterization and detection results.

In this paper, we propose an economic model predictive control (EMPC) strategy for the constrained optimization problem of the wave energy converter system. Unlike standard model predictive control (MPC) that uses the tracking cost function, in EMPC, a general economic cost function that directly reflects the economic objective of the system is established. In the wave energy converter (WEC), the general economic cost function aims to maximize the energy extracted from ocean waves and minimize the operation cost. The terminal equality constraint in the optimization problem is used to steer the system state to the steady-state at the end of the optimization horizon. Then the controller solves the non-convex optimization problem in real-time. The auxiliary optimization problem is used for the stability analysis, and the convergence of the system can be proved via the Lyapunov technique. Several simulation results are presented to demonstrate the effectiveness of the proposed strategy.

The generation unit aggregation is frequently used in a virtual power plant which participates in real-time spot markets. In this letter, an analytical method is proposed to quickly and precisely compute a piecewise quadratic function of aggregated units. Three propositions are introduced and proven in the letter in order to guarantee the accuracy of the equivalent model after aggregation. Numerical results verify the validity of the proposed method.